A novel radar target recognition method for open and imbalanced high-resolution range profile

• Published in: Digital signal processing 2021-11, Vol.118, p.103212, Article 103212
• Main Authors: Zhang, Xianwen, Wang, Wenying, Zheng, Xuanxuan, Wei, Yao
• Format: Article
• Language: English
• Subjects: High-resolution range profile (HRRP); Imbalanced learning; Open set recognition; Radar automatic target recognition (RART)
• ISSN: 1051-2004; 1095-4333
• DOI: 10.1016/j.dsp.2021.103212

•Introduce a more realistic radar ATR task as open and imbalanced HRRP recognition.•Design a novel neural network consists of memory, angular penalty and dual-attention.•This method can recognize targets accurately, outperforming traditional ways.•Dual-attention provides strong discriminative feature representation.•Arcloss and memory optimize the intra-class similarity and inter-class variability. High-resolution range profile (HRRP) data often has imbalanced and open-ended distribution in realistic radar target recognition task, requiring the recognition system classify targets among majority and minority classes and detect unseen targets accurately. In this paper, we propose a novel algorithm for open and imbalanced HRPP recognition tasks, by learning from realistic data distribution and optimizing the accuracy over the majority, minority and open classes. This method maps the HRRP data to the latent feature space, enhances the feature by sharing knowledge from majority to minority class, and provides a scalar indicating the familiarity to known classes. Dual-attention is developed to provide strong discriminative feature representation. An angular penalty is employed in the loss function to optimize the intra-class similarity and inter-class variability. Experiments on measured data prove that the proposed algorithm outperforms other existing methods under both close and open settings with accuracy increased significantly, and more discriminative representation. This study provides a promising and effective approach for open and imbalanced HRRP target recognition.